RESEARCH | SYNTHETIC YEAST GENOME

◥ ing 10 codons near the 3′ end of the PRE4 open RESEARCH ARTICLE SUMMARY reading frame (ORF) caused a ~twofold reduc- tion in Pre4 protein level without affecting RNA abundance. Reverting the codons to the WT se- SYNTHETIC BIOLOGY quence corrected both the Pre4 protein level and the phenotype. We hypothesize that the forma- Synthesis, debugging, and effects of tion of a stem loop involving recoded codons

◥ underlies reduced Pre4 pro- synthetic chromosome consolidation: ON OUR WEBSITE tein level. Read the full article Sc2.0 chromosomes (synI at http://dx.doi. to synXVI) are constructed synVI and beyond org/10.1126/ individually in discrete science.aaf4831 strains and consolidated ...... Leslie A. Mitchell, Ann Wang, Giovanni Stracquadanio, Zheng Kuang, Xuya Wang, into poly-synthetic (poly- Kun Yang, Sarah Richardson, J. Andrew Martin, Yu Zhao, Roy Walker, Yisha Luo, syn) strains by “endoreduplication intercross.” Hongjiu Dai, Kang Dong, Zuojian Tang, Yanling Yang, Yizhi Cai, Adriana Heguy, Consolidation of synVI with synthetic chromo- Beatrix Ueberheide, David Fenyö, Junbiao Dai, Joel S. Bader, Jef D. Boeke* somes III (synIII) and IXR (synIXR) yields a triple-synthetic (triple-syn) strain that is ~6% INTRODUCTION: Total synthesis of designer some consolidation to uncover possible synthetic synthetic overall—with almost 70 kb deleted, in- chromosomes and genomes is a new paradigm genetic interactions and/or perturbations of native cluding 20 tRNAs, and more than 12 kb recoded. for the study of genetics and biological systems. cellular networks as the number of designer changes Genome sequencing of double-synthetic (synIII The Sc2.0 project is building a designer yeast increases is the next major step for the Sc2.0 project. synVI, synIII synIXR, synVI synIXR) and triple- genome from scratch to test and extend the lim- syn (synIII synVI synIXR) cells indicates that its of our biological knowledge. Here we describe RESULTS: SynVI was rapidly assembled using suppressor mutations are not required to en- the design, rapid assembly, and characterization nine sequential steps of SwAP-In (switching auxo- able coexistence of Sc2.0 chromosomes. Pheno- of synthetic chromosome VI (synVI). Further, we trophies progressively by integration), yielding a typic analysis revealed a slightly slower growth on March 14, 2017 investigate the phenotypic, transcriptomic, and ~240-kb synthetic chromosome designed to rate for the triple-syn strain only; the combined effect of tRNA deletions on different chromosomes might underlie this result. Transcriptome and pro- teomeanalysesindicatethatcellularnetworksare largely unperturbed by the existence of multiple synthetic chromosomes in a single cell. However, a second bug on synVI was discovered through proteomic analysis and is associated with alteration of the HIS2 transcription start as a consequence of tRNA deletion and loxPsym site insertion. De-

spite extensive genetic alterations across 6% of http://science.sciencemag.org/ the genome, no major global changes were de- tected in the poly-syn strain “omics” analyses.

CONCLUSION: Analyses of phenotypes, tran- scriptomics, and proteomics of synVI and poly- syn strains reveal, in general, WT cell properties andtheexistenceofrarebugsresultingfromge- nome editing. Deletion of subtelomeres can lead Downloaded from to gene silencing, recoding deep within an ORF Debugging synVI and characterization of poly-synthetic yeast cells. (A) The second Sc2.0 can yield a translational defect, and deletion of ele- chromosome to be constructed, synVI, encodes a “bug” that causes a variable colony size, ments such as tRNA genes can lead to a complex dubbed a “glycerol-negative growth-suppression defect.” (B) Synonymous changes in the essential transcriptional output. These results underscore PRE4 ORF lead to a reduced protein level, which underlies the growth defect. (C) The poly-synthetic the complementarity of transcriptomics and pro- strain synIII synVI synIXR directs growth of yeast cells to near WT fitness levels. teomics to identify bugs, the consequences of designer changes in Sc2.0 chromosomes. The proteomic consequences associated with consoli- Sc2.0 specifications. We observed partial silenc- consolidation of Sc2.0 designer chromosomes dation of three synthetic chromosomes—synVI, ing of the left- and rightmost genes on synVI, into a single strain appears to be exceptionally synIII, and synIXR—into a single poly-synthet- each newly positioned subtelomerically relative well tolerated by yeast. A predictable excep- ic strain. to their locations on native VI. This result sug- tion to this is the deletion of tRNAs, which will be gests that consensus core X elements of Sc2.0 restored on a separate neochromosome to avoid RATIONALE: A host of Sc2.0 chromosomes, universal telomere caps are insufficient to fully synthetic lethal genetic interactions between including synVI, have now been constructed buffer telomere position effects. The synVI strain deleted tRNA genes as additional synthetic chro- in discrete strains. With debugging steps, where displayed a growth defect characterized by an mosomes are introduced.▪ thenumberofbugsscaleswithchromosome increased frequency of glycerol-negative colonies. length, all individual synthetic chromosomes have The defect mapped to a synVI design feature in The list of author affiliations is available in the full article online. PRE4 YFR050C *Corresponding author. Email: [email protected] been shown to power yeast cells to near wild-type the essential gene ( ), encoding Cite this article as L. A. Mitchell et al., Science 355,eaaf4831 (WT) fitness. Testing the effects of Sc2.0 chromo- the b7 subunit of the 20S proteasome. Recod- (2017). DOI: 10.1126/science.aaf4831

Mitchell et al., Science 355, 1045 (2017) 10 March 2017 1of1 RESEARCH | SYNTHETIC YEAST GENOME

◥ on synVI were de novo as a result of incorpora- RESEARCH ARTICLE tion (5), corresponding to a mutation frequency of ~1 × 10−5, similar to that observed for synIII (6). Two nonsynonymous mutations affected the SYNTHETIC BIOLOGY essential MOB2 gene (YFL034C-B) and conferred a notable fitness defect (fig. S3); their correction to the designed sequence substantially improved Synthesis, debugging, and effects of growth (fig. S4) (5). RNA sequence (RNA-seq) profiling of synVI strains [yLM175 (yeast_chr06_9_01) and yLM953 synthetic chromosome consolidation: (yeast_chr06_9_03)] revealed only a small num- ber of significant changes in gene expression synVI and beyond relative to the wild type (Fig. 1C and fig. S5). The two significantly down-regulated genes on Leslie A. Mitchell,1,2 Ann Wang,3,4 Giovanni Stracquadanio,3,5,6 Zheng Kuang,1,2 synVI each lie adjacent to a terminal universal Xuya Wang,1,2 Kun Yang,3,5 Sarah Richardson,3,5 J. Andrew Martin,1,2 Yu Zhao,1,2 telomere cap (UTC) (6): YFL055W encoding Agp3, Roy Walker,7 Yisha Luo,7 Hongjiu Dai,8 Kang Dong,8 Zuojian Tang,1,2 Yanling Yang,9 a low-affinity amino acid permease (7), and Yizhi Cai,7 Adriana Heguy,10 Beatrix Ueberheide,1,9 David Fenyö,1,2 Junbiao Dai,4 YFR055W encoding Irc7, a beta-lyase involved Joel S. Bader,3,4 Jef D. Boeke1,2* in thiol production (8). To test for gene silencing by telomere position effect (TPE) due to deleting We describe design, rapid assembly, and characterization of synthetic yeast Sc2.0 native subtelomeres (Fig. 1C, inset), we gen- chromosome VI (synVI). A mitochondrial defect in the synVI strain mapped to synonymous erated circularized derivatives of native VI and coding changes within PRE4 (YFR050C), encoding an essential proteasome subunit; synVI (Fig. 1B) (5). Expression of YFL055W and Sc2.0 coding changes reduced Pre4 protein accumulation by half. Completing Sc2.0 YFR055W from ring synVI was similar to levels specifies consolidation of 16 synthetic chromosomes into a single strain. We investigated observed in the native linear VI strain (Fig. 1D). phenotypic, transcriptional, and proteomewide consequences of Sc2.0 chromosome Native VI subtelomeres encode core X elements, consolidation in poly-synthetic strains. Another “bug” was discovered through proteomic and the left subtelomere contains a Y′ element. on March 14, 2017 analysis, associated with alteration of the HIS2 transcription start due to transfer RNA Although the core X sequence clearly has some deletion and loxPsym site insertion. Despite extensive genetic alterations across 6% insulator-like function (9), our data indicate that of the genome, no major global changes were detected in the poly-synthetic strain consensus core X elements of the UTC are in- “omics” analyses. This work sets the stage for completion of a designer, synthetic sufficient to fully insulate telomere-proximal eukaryotic genome. promoters from TPE and suggests that one func- tion of long subtelomeric regions is to fully buffer the transcriptome against silencing. otal synthesis of designer chromosomes and five spliceosomal introns in native chromosome After correcting nonsynonymous mutations in genomesisanewparadigmforthestudyof VI genes were deleted, a single specialized non- MOB2, the synVI strain [yLM402 (yeast_chr06_ genetics and biological systems. The Sc2.0 spliceosomal intron, encoded by HAC1 (YFL031W), 9_02)] still had a noticeable growth defect or project is building a designer yeast genome was retained because of its critical role in regulation “bug” characterized by variable colony size, with

T http://science.sciencemag.org/ from scratch to test and extend the limits of the unfolded protein response (3, 4). Retaining roughlyhalfofallcoloniesgrowingtoasmaller of our biological knowledge. this intron remains consistent with an overall Sc2.0 overall size. The small colonies were respiratory- project goal to remove all spliceosomal introns deficient, as they could not grow on medium with — SynVI Design, assembly, and and eventually test whether the spliceosome has glycerol as the sole carbon source. Analysis of the debugging biological functions beyond removing introns from mitochondrial DNA (mtDNA) of the glycerol- Here we report the construction of the second pre-mRNAs. The final synVI designed chromo- negative synVI colonies revealed that segments full-length Sc2.0 chromosome, synVI, designed to some is 242,745 base pairs (bp) in length, 11.3% of the 15S rRNA, ATP6, and COX2 genes were Sc2.0 specifications using the genome-editing suite shorter than native VI. lost at high frequency (fig. S6). Glycerol-negative BioStudio (Fig. 1A and fig. S1) (1, 2). Although all SynVI was efficiently assembled using SwAP- synVI colonies were surprisingly large relative to Downloaded from In (switching auxotrophies progressively by inte- glycerol-negative derivatives of wild-type (WT)

1 gration) (1, 2), enabled by segmentation of the cells. This “glycerol-negative growth-suppression Department of Biochemistry and Molecular Pharmacology, ” New York University Langone School of Medicine, New York, designed chromosome into 9 megachunks (each defect is best visualized by comparing the colony NY 10016, USA. 2Institute for Systems Genetics, New York of 30 to 40 kb) and 26 chunks (~10-kb each) size of WT and synVI cells pretreated in ethidium University Langone School of Medicine, New York, NY 10016, (Fig. 1A). The production of 10-kb chunks was bromide, which converts cells to r0 by inducing 3 USA. High Throughput Biology Center, Johns Hopkins performed using standard gene-assembly meth- thelossofmtDNA(Fig.1E).Thedefectdidnot University School of Medicine, Baltimore, MD 21205, USA. 4Key Laboratory for Industrial Biocatalysis (Ministry of ods, and nearly all chunks were successfully built depend on nonsynonymous synVI mutations (table Education), Key Laboratory of Bioinformatics (Ministry of as full-length, sequence-verified constructs. In S1 and fig. S7). We also evaluated the fitness of Education), Center for Synthetic and Systems Biology, nine sequential steps, we incorporated more than the synVI strain by plating serial dilutions on a School of Life Sciences, Tsinghua University, Beijing 100084, 240 kb of synthetic DNA into a haploid yeast strain diverse panel of media and growth conditions (6) China. 5Department of Biomedical Engineering and Institute of Genetic Medicine, Whiting School of Engineering, Johns to build synVI (Fig. 1B). At each integration step, and found no condition that substantially af- Hopkins University, Baltimore, MD 21218, USA. 6School of clones that exclusively produced synthetic poly- fected growth (fig. S8). Computer Science and Electronic Engineering, University of merase chain reaction tag (PCRTag) amplicons, We mapped the defect to the gene PRE4 (YFR050C) 7 Essex, Wivenhoe Park, Colchester CO4 3SQ, UK. Center for used as watermarks for the synthetic sequence using a meiotic recombination strategy (Fig. 1, F Synthetic and Systems Biology, School of Biological Sciences, University of Edinburgh, Edinburgh EH9 3JL, UK. (2), were identified (fig. S2). DNA sequencing of andG,andfigs.S9toS13)(5). PRE4 encodes the 8GenScript, Piscataway, NJ 08854, USA. 9Proteomics the synVI genome revealed 10 point mutations essential b7 subunit of the 20S proteasome (10). Resource Center, Office of Collaborative Science, New York on synVI compared with the designed sequence. The Pre4 C terminus plays a structural role in 20S University Langone School of Medicine, New York, NY 10016, All point mutations were in open reading frames, proteasome assembly (11)whiletheN-terminal USA. 10Genome Technology Center, New York University Langone School of Medicine, New York, NY 10016, USA. and five yielded nonsynonymous mutations (table eight amino acids are processed (12)uponassem- *Corresponding author. Email: [email protected] S1). Only 2 of the 10 point mutations recovered bly (13). To test whether a reduced PRE4 copy

Mitchell et al., Science 355, eaaf4831 (2017) 10 March 2017 1of11 RESEARCH | RESEARCH ARTICLE | SYNTHETIC YEAST GENOME number might underlie the glycerol-negative partial rescue, and replacing four consecutive as WT.PRE4, irrespective of chromosomal con- growth-suppression defect, we made diploids rare codons (R2, S3, S4, R5) with the WT ver- text (native VI or synVI). No protein products of heterozygous for PRE4 in VI/VI or synVI/synVI sions (WT.RSSR) in synthetic PRE4 on synVI unexpected molecular weight were observed (Fig. strains. The VI/VI pre4D/PRE4 strain displayed rescued the defect (fig. S14). Conversely, intro- 2E). PCRTag recoding, leading to protein-related a phenotype very similar to that of the PRE4/ ducing the four rare codons (SYN.RSSR)into defects, could result from a ribosomal stall, un- PRE4 synVI/synVI strain (Fig. 2A). The designer WT PRE4 on native VI was insufficient to pheno- derpinned by generation or destruction of RNA features encoded by the SYN.PRE4 allele include copy the defect (fig. S14). modification sites, formation of previously uni- a TAG/TAA stop-codon swap and a PCRTag pair Genetic analyses suggested that PRE4 was dentified mRNA secondary structure, or an im- with recoded sequences encompassing 10 codons underexpressed in synVI. Given that the PRE4 balance between tRNA supply and demand. Any each (Fig. 2, B and C). Installing the WT.PRE4 al- transcript level was normal (Fig. 1C and fig. S5) of these factors could reduce protein accumula- lele into synVI or the SYN.PRE4 allele into native (5), we examined Pre4 protein accumulation. Hap- tion by reducing translational output or causing VI rescued or phenocopied the glycerol-negative loid strains expressing either synthetic (SYN.PRE4) misfolding coupled to protein degradation. In growth-suppression defect, respectively (Fig. or WT (WT.PRE4) alleles with N-terminal HA tags the synVI strain, 10 tRNA genes were deleted 2D). The defect was further tracked to the syn- were constructed; each was integrated into the andmorethan9000bpwererecoded(tableS2). thetic PCRTag closer to the PRE4 3′ end (Fig. WT (VI) or synthetic (synVI) PRE4 chromosomal However, WT cells expressing the WT-SYN.PRE4 2D). We also introduced individual WT codons locus. Although the Pre4 N terminus is cleaved allele in the native VI background phenocopied into an otherwise synthetic PRE4 allele in the (12), we tagged this end rather than the C termi- the glycerol-negative growth-suppression defect, synVI background and found that no single WT nus to capture potential translational frameshifts which suggests a mechanism independent of codon could rescue the glycerol-negative growth- predicted to result in changes in protein length. tRNA supply and demand. Consistent with this, suppression defect entirely (fig. S14). However, We found that strains encoding SYN.PRE4 ex- episomal expression of the 10 tRNAs deleted from WT codons at positions S4, R5, and N6 showed pressed about half as much Pre4 protein product synVI had no effect on Pre4 protein level (fig. S14).

Fig. 1. SynVI design, assembly, and bug mapping. (A)The~242-kb synthetic chromosome VI (synVI) encodes all designer features of on March 14, 2017 Sc2.0 chromosomes (1, 2, 6). Deleted elements are denoted with the letter “x.” SynVI is segmented into nine megachunks (A to I) for SwAP-In (switching auxotrophies progressively by integration). Gene order and orientation are indicated (red, essential; purple, required for fast growth; dark blue, nonessential with function characterized; light blue, nonessential with uncharac- terized function; gray, dubious). http://science.sciencemag.org/ UTC, universal telomere cap; ARS, autonomously replicating sequence; CEN, centromere. (B) SynVI (~242 kb) (yLM402) migrates faster than native VI (~270 kb) (BY4741) by pulsed-field gel electrophoresis. Circular deriva- tives of native VI (yLM717) and Downloaded from synVI (yLM728) do not penetrate the gel (black arrowheads). (C) RNA-seq data of synVI (yLM953) represented as a volcano plot. Genes on synVI are shown in red. (Inset) Schematic of synVI telo- meres along with right- and leftmost genes. Sc2.0 UTCs encode consen- sus core X elements flanked by loxPsym sites (green diamonds) with flanking yeast telomere repeat sequence [(C1-3A)~115); (TG1-3)~115)]. (D) Relative expression of AGP3 and IRC7 from strains carrying native VI or synVI in linear (BY4741, yLM402) or ring structures (yLM718, yLM728). Error bars indicate SD. (E) The synVI strain (yLM402) produces glycerol-negative colonies that grow faster on yeast extract peptone dextrose (YPD) medium than native VI. Spontaneously generated glycerol-negative colonies (left) compared with r0 cells generated by pretreatment in ethidium bromide (EtBr) (right). (F) Meiotic PCRTag mapping of two full tetrads (10 and 13, A to D) (backcross of synVI strain yLM402 to WT strain BY4742), with 15 synthetic PCRTags spanning synVI from left to right. Native VI is inferred by the absence of amplicons. Red lettering denotes spores with growth defects, as shown by single-colony growth. Also see figs. S9 and S10. (G) An extra copy of synthetic YFR050C rescues the synVI growth defect (yLM707, yLM875-yLM878).

Mitchell et al., Science 355, eaaf4831 (2017) 10 March 2017 2of11 RESEARCH | RESEARCH ARTICLE | SYNTHETIC YEAST GENOME

In silico mRNA secondary structure prediction of than 12 kb recoded (table S2). Genome sequencing thetic chromosomes are consolidated, the deleted sequence spanning the synthetic PCRTag indicates of the four poly-syn strains uncovered only a small tRNAs will be restored on a separate neochromo- potential formation of a stem loop mediated by number of point mutations relative to design (table some. Doubling time, colony size, and morphology recoded bases (S4 and R5 codons pairing with L9 S3), all of which were common to all sequenced were generally indistinguishable from the wild and A10) (Fig. 2F). Involvement of S4 and R5 in a strains, indicating that they existed before endo- type for all double-synthetic (double-syn) strains stem loop is consistent with our genetic data (fig. reduplication intercrossing. This result suggests that on all media types and conditions tested (Fig. S14) and, taken together with ruling out tRNA new suppressor mutations need not arise to enable 3A and fig. S15). supply and demand, implicates secondary structure coexistence of multiple synthetic chromosomes. RNA sequencing of poly-syn strains revealed of the SYN.PRE4 mRNA transcript as mechanis- To further evaluate the effects of consolidat- only a small number of significant transcriptome tically underlying reduced Pre4 protein level. The ing synthetic chromosomes, we undertook pheno- changes in the poly-syn strains (Fig. 3C and fig. connection between Pre4 level and the glycerol- typic, transcriptomic, and proteomic profiling of S16) (5). Surprisingly, in the triple-syn strain, ex- negative growth-suppression defect remains to poly-syn strains. We compared the growth of poly- pression of the four genes encoded by the native be explored. syn strains to that of WT strains and observed a 2-micron plasmid (FLP1, REP1, REP2, RAF1)(14) ~15% increase in doubling time of the triple-syn was significantly reduced (Fig. 3C). From genome Effects of consolidation strain in a multiwell-plate liquid culture growth sequence data, the depth of reads mapping to the of synthetic chromosomes assay (Fig. 3A). However, overall colony size and 2-micron plasmid was virtually nonexistent, sug- The Sc2.0 chromosomes (synI to synXVI) are morphology on both yeast extract peptone dex- gestingthatthe2-micronplasmidhadnearlybeen being constructed individually in discrete strains trose (YPD) and 3% glycerol plates at 30° and lost from the triple-syn strain at the time of se- (1, 2, 6). Using “endoreduplication intercross,” poly- 37°C were comparable after 7 days (Fig. 3B). In quencing. Curiously, outright loss of the 2-micron synthetic (poly-syn) chromosome strains encoding contrast,colonysizeonsyntheticcompleteme- plasmid was not observed in any of the double-syn two (synIII synVI; synIII synIXR; synVI synIII) and dium was slightly smaller for triple-syn cells than chromosome strains, although it was absent in three synthetic chromosomes (synIII synVI synIXR) for the wild type (Fig. 3B). One possible cause an early version of the synX strain (15). Endoredupli- were constructed (1). The triple-synthetic strain underlying reduced growth of the synIII synVI cation backcross of synX led to recovery of the (triple-syn) is ~6% synthetic overall, with almost synIXR strain is the deletion of 20 tRNAs (table 2-micron plasmid, which suggests that its loss is 70 kb deleted (including 20 tRNAs) and more S2). Importantly, as increasing numbers of syn- not a function of the synthetic chromosomes. on March 14, 2017 http://science.sciencemag.org/ Downloaded from

Fig. 2. PRE4 PCRTag recoding underlies synVI glycerol-negative growth- synthetic PRE4 allele (SYN.PRE4). The 5′ PCRTag (black bar) has no impact. suppression defect. (A) PRE4 haploinsufficiency. Deleting one copy of PRE4 (Strains, top row: BY4741, yLM849, yLM853, yLM851; bottom row: yLM949, (WT/WT WT.PRE4/wt.pre4D; yLM770) in the wild-type (WT) diploid (VI/VI; BY4743) yLM402, yLM957, yLM953.) (E) Pre4 protein level. N-terminally HA-tagged Pre4, phenocopies the glycerol-negative growth-suppression defect observed in the expressed from the PRE4 locus on synVI or native VI as a synthetic or WT allele. synVI diploid (synVI/synVI; yLM544) and synVI SYN.PRE4 heterozygous Protein level is expressed as a percentage of the wild type. (Strains, from left to diploids (synVI/synVI PRE4/pre4D; yLM772). Glycerol-negative colony fre- right: BY4741, yLM883, yLM882, yLM867, yLM866, yLM1212, yLM1214, yLM1216, quency as percentage of total colonies is indicated. (B)SyntheticYFR050C/ yLM1218.) (F) mFold secondary structure predictions for SYN.PRE4 and WT. PRE4 locus. ORF, open reading frame. (C) PRE4 WT and synthetic (SYN) PRE4 alleles. Predictions were generated using the 20 bases of the 3′ PRE4 PCRTag sequences are synonymous. Codon usage is indicated as frequency per PCRTag plus 6 flanking bases using the mFold Web server (31). 3′ PRE4 PCRTag thousand codons. (D) Installing the 3′ synthetic PCRTag (green bar) provokes codons S4, R5, A10, and L9 are indicated in red boxes for the synthetic (left, the defect in WT cells (VI) and rescues the defect in synVI cells. PCRTags in the SYN.PRE4) and WT (right, WT.PRE4) alleles. DG, change in Gibbs free energy.

Mitchell et al., Science 355, eaaf4831 (2017) 10 March 2017 3of11 RESEARCH | RESEARCH ARTICLE | SYNTHETIC YEAST GENOME

Rather,itislikelythatthelossofthe2-micron used tandem mass tag labeling to quantitatively triple-syn strain (Fig. 3D and table S4). We di- plasmid is a consequence of multiple rounds of evaluate the triple-syn proteome relative to the rectly tested six of these proteins by immunoblot genetic manipulation and single-colony purification. wild type. The SYN-WT.PRE4 allele was encoded analysis and confirmed altered expression only PRE4 recoding reduced protein expression. in the triple-syn strain, so we did not expect al- for His2, encoded by synVI (Fig. 3E and fig. S17). To test for more widespread proteome changes tered levels of Pre4 protein. Of 3122 proteins We found that His2 protein level depended on in the triple-syn strain, mediated by recoding whose relative abundance was determined, only presence of an upstream tRNA gene [tA(AGC)F] events and/or tRNA supply and demand, we 11 displayed significantly altered levels in the and that its replacement with a loxPsym site on March 14, 2017 http://science.sciencemag.org/ Downloaded from

Fig. 3. Poly-synthetic strain characterization. (A) Growth curve and doubling analysis of triple-syn cells (SYN) (yLM896) compared with WT cells (BY4742). time of synthetic strains (30°C, YPD, 24 hours). (Strains: BY4741, yLM422, Proteins with significantly changed expression are shown in red and named. yLM953, yLM461, yLM890, yLM758, yLM892, yLM896.) Error bars indicate (E) Immunoblot evaluating expression of tagged SYN or WT HIS2 (HIS2-HA) SD. OD, optical density. (B) Colony size and morphology of the triple- from either the native VI (WT) or synVI (SYN) chromosome. Inserted up- synthetic (triple-syn) strain (yLM896) compared with the wild type (BY4742). stream of HIS2 is either the native tRNA gene [tA(AGC)F] or a loxPsym site. (C) RNA-seq data of the triple-syn strain (yLM896) represented as a volcano Intervening lanes are cropped. (Strains, from left to right: BY4741, yLM953, plot. Genes are categorized according to color as indicated. The triple-syn yLM1198 to yLM1205.) (F) The synthetic HIS2 locus. (G)RNA-seqreads strain (MATa MET15, lys2D0, ura3) transcriptome was determined relative to mapping to the SYN.HIS2 and WT.HIS2 alleles from the triple-syn and WT that of a WT strain (BY4741; MATa, met15D, LYS2, ura3D0). Genotypic dif- strains. The position of the tRNA gene [tA(AGC)F]innativeVIandtheloxPsym ferences underlie most differentially expressed genes. (D)Whole-proteome site replacing it in the synVI chromosome are indicated.

Mitchell et al., Science 355, eaaf4831 (2017) 10 March 2017 4of11 RESEARCH | RESEARCH ARTICLE | SYNTHETIC YEAST GENOME decreased cellular His2 levels (Fig. 3, E and F). purified chunk fragments were first mixed to- no marker and was designed to overwrite a pre- In the triple-syn RNA-seq data set, we discovered gether and precipitated using ethanol (3 volumes, existing URA3 marker in synVI.H. For this reason a set of reads mapping upstream of the HIS2 100%) and 3 M sodium acetate pH 5.2 (1/10 vol- selection for the full-length synVI chromosome startsitethatwereabsentfromthewildtype(Fig. ume). Instead of equimolar quantities, an ever- was achieved on 5-FOA (5-fluoro-orotic acid) and 3G). Replacement of the tRNA with a loxPsym decreasing amount of DNA of each chunk in the yieldedanunmarkedchromosome. Yeast trans- site may create a cryptic start site, such that left to right direction was used to maximize formations were carried out using a standard not all mRNA yields His2 translation product the percentage of correct ligation products con- lithium acetate/polyethylene glycol protocol (17), and/or that transcription from the upstream taining the selectable marker (e.g., 500 ng E1: except cells were heat shocked for only 15 min in region might interfere with expression from 200ngE2:100ngE3). the presence of 10% DMSO at 42°C and prior to the native His2 promoter (Fig. 3G). This result Two chunks, A1 and B3, were built as partial, plating were incubated in 400 ml 5 mM CaCl2 for underscores the importance of proteomic anal- nonoverlapping subconstructs that had been 10 min at room temperature. In all cases the en- ysis as a complementary tool to RNA-seq to iden- sequence-verified. For megachunk A, the two tire in vitro chunk ligation product was trans- tify consequences of designer changes in Sc2.0 A1 “minichunks” (A1.1 and A1.2) were pre-joined formed into yeast. Strains used for integration chromosomes. In total, quantitative proteomic in an equimolar in vitro ligation reaction using (from left to right): yLM097 (yfl054CD::kanMX), analysis allowed us to evaluate relative expres- a PstI site and gel-purified prior to a second yJS272 (synVI.A), yLM067 (synVI.B), yLM068 sionlevelof~50%ofproteinsencodedbychro- round of in vitro ligation with A2 and A3 (2). (synVI.C), yLM085 (synVI.D), yLM086 (synVI.E), mosomes synIII, synVI, and synIXR. The paucity The DNA corresponding to chunk B3 was de- yLM088 (synVI.F), yLM090 (synVI.G), yLM091 of changes indicates minimal impact on the livered in three separate constructs (B3.1, B3.2, (synVI.H). proteome from designer changes to the genome B3.3). We designed primers with ~40-bp over- sequence. hangs to generate overlap between these mini- Pulsed-field gel electrophoresis chunks as well as a vector sequence [pRS413 (16)] Full-length yeast chromosomes were prepared Conclusions capable of single-copy replication in S. cerevi- in agarose plugs as described (18). Chromosomes Phenotypic, transcriptomic, and proteomic anal- siae.ThethreeB3PCRfragmentsplustheSmaI- were separated by clamped homogeneous electric ysis of the synVI and poly-syn strains reveals, linearized pRS413 vector were cotransformed field (CHEF) gel electrophoresis using the CHEF- in general, a WT set of properties, with at least into the WT yeast strain BY4741 for “in yeasto DR III Pulsed-Field Electrophoresis System (Bio- three unexpected “bugs” resulting from genome assembly” by homologous recombination. A con- Rad) using the auto algorithm function to separate editing. Deletion of subtelomeres led to synVI struct encoding a full-length B3 insert was re- chromosomes ranging in size from 200 to 500 kb on March 14, 2017 terminal gene silencing, recoding deep within covered into E. coli from a single yeast colony [6 V/cm, switch time 24:03 to 44:69 s, run time PRE4 resulted in a translational defect, and de- (pAW006). The insert size was verified by re- 34:02 hours, 14°C, 0.5X Tris-Borate-EDTA buffer, letion of an upstream tRNA gene caused a trans- striction digest and the presence of all synthetic 1% gel prepared with low melting point agarose lational phenotype associated with activation of PCRTags confirmed by PCR. However, the se- (Lonza, 50100)]. Gels were stained with 5 mg/ml a cryptic start site and/or promoter interference. quence of the B3 insert was not verified prior to ethidium bromide in water postelectrophoresis, Consolidation of Sc2.0 designer chromosomes incorporation into yeast as part of megachunk destained in water, and then imaged. into a single strain appears to be exceptionally B. After whole genome sequencing of the synVI well tolerated by yeast, despite the large num- yeast strain (yLM093) we discovered seven mu- PCRTag analysis ber of designer changes. As additional synthetic tations encoded within the B3 segment (table PCRTag analysis was performed in 2.5-mlreac- chromosomes are introduced, deleted tRNA genes S2). Subsequent Sanger sequencing of the B3 tion volumes using 2X GoTaq Hot Start Color- will be restored on a separate “neochromosome” assembled chunk clone revealed that six of the less Master Mix (Promega, M5132). Master Mix

to avoid synthetic genetic interactions between mutations preexisted in B3 but not in B3.1, (1X) was arrayed into each well of a 384 well http://science.sciencemag.org/ tRNA gene deletions. B3.2, or B3.3, suggesting they arose during the PCR plate (Greiner, 785201) and the acoustic dis- chunk assembly process, underscoring the im- penser Echo 550 (LabCyte) was used to transfer Materials and methods portance of sequence-verifying chunks prior to genomic DNA [5 nl, prepared as previously de- Chunk synthesis and in vitro incorporation into yeast. scribed (19)] and primers (20 nl, 30 mM premixed megachunk assembly For megachunk D, an internal AvaI site was forward and reverse) into the appropriate wells. With a few exceptions (chunks A1 and B3, see identified at the junction between the synthetic PCR products were analyzed by 1% gel elec- below) synthetic DNA constructs were deliv- DNA and the URA3 selectable marker in chunk trophoresis in 1X Tris-Taurine-EDTA or on the ered as subcloned and sequence-verified ~10 kb D3. This was problematic as AvaI had been as- LabChip GXII (Perkin-Elmer). Downloaded from chunks. All chunk constructs were transformed signed as the restriction enzyme site for excising During integration of megachunk D we re- into E. coli and grown at either 25°C (chunk C1) chunk D3 at its 5′ end. To overcome this prob- peatedly recovered clones that produced both or 30°C (all other constructs). Top10 cells were lem, restricted D1, D2, and D3 (missing the synthetic and WT amplicons at the YFL009W.2 used to propagate all chunk constructs except URA3 sequence and 3′ terminal homology to locus (fig. S2A). We hypothesized the WT PCRTag F1 and F3, which were transformed in Copy- the native chromosome following digestion with primers were annealing to a secondary locus in Cutter EPI-400 cells (Epicenter, C400CH10) to AvaI) fragments were joined by in vitro ligation thegenome.TheWTYFL009W.2 amplicon se- maintain low copy number prior to the addition and cotransformed into yeast with a PCR- quence was used in a BLAST search against the of Induction Solution (Epicentre, CIS40025) to generated URA3 marker derived from the D3 native yeast genome, returning two hits: its own the liquid growth medium as per the manu- chunk with ~1-kb terminal homology to the native location on chromosome six, plus a nearly iden- facturer’s instructions. Constructs were isolated chromosome. tical region near YER066W on chromosome 5 from saturated bacterial cultures by standard (chrV 289709-290199) (fig. S2B). We deleted the alkaline lysis followed by isopropanol precipita- Integrating megachunks by SwAP-In chrV region with a KanMX cassette in the synVI tion. ~5 mg was digested with the appropriate SynVI was built from left to right in nine se- strain to generate yLM096 and repeated the restriction enzymes and the synthetic DNA frag- quential integrative transformations, each in- PCRTag analysis. In the absence of the chrV locus ments gel-purified using the Zymoclean Large volving a “megachunk” of synthetic DNA (A-I) the WT YFL009W.2 PCRTag primers no longer Fragment DNA Recovery Kit (Zymo Research, segmented into three or four ~10-kb pieces ar- produced an amplicon (fig. S2C). D4045). Megachunks were then assembled by bitrarily called “chunks.” The rightmost chunk in vitro ligation overnight at 16°C in 10-mlreaction of each megachunk encodes an auxotrophic se- Genome sequencing volumes using 0.1X T4 DNALigase(Enzymatics, lectable marker, either LEU2 or URA3.Thesin- Genomic DNA for sequencing was prepared as L6030-HC-L). For ligation, the appropriate gel- gle exception is megachunk I, which encodes described previously (6). A whole genome shotgun

Mitchell et al., Science 355, eaaf4831 (2017) 10 March 2017 5of11 RESEARCH | RESEARCH ARTICLE | SYNTHETIC YEAST GENOME library was prepared as follows: 500 ng of genomic of HAC1 is apparently specific to the early version ment. First strand cDNA was prepared using the DNA were sheared to 500 bp fragments using a of synVI through an undetermined mechanism. SuperScript III First-Strand Synthesis SuperMix Covaris E220 ultrasonicator, and a library was (Invitrogen, 18080-400). The expression level of prepared using the Kapa Low-Throughput “With Poly-syn strains genes of interest was tested using SsoAdvanced bead” Library Preparation Kit Standard (Kapa The transcriptome of four poly-syn strains SYBR Green Supermix (BioRad, 172-5265) in a Biosystems, KK8231) without any PCR amplifica- (yLM890 synIII synVI; yLM758 synIII synIXR; 4-ml reaction volume with technical quadrupli- tion, and sequenced as 50 base paired-end reads yLM892 synVI synIXR) was determined by RNA- cates. Here, reactions were assembled using an on an Illumina HiSeq 2500 (v4 chemistry) ma- seq. For this experiment, BY4741 was used as the acoustic droplet ejection robot (Echo 550, LabCyte) chine. Sequence analysis was carried out as previ- WT reference strain. BY4741 is a MATa strain to transfer primers (50 nl of 50 mMeachpre- ously described (6). and many of the poly-syn strains are MATa,spe- mixed; final concentration 0.625 mM each) and cifically synIII synVI (yLM890), synIII synIXR cDNA (75 nl) into mastermix pre-arrayed in a SynVI (yLM758), synIII synVI synIXR (yLM896) (table 384-well plate (Greiner, 785201). qPCR was per- The synVI genome sequence was first deter- S5). Thus, in the RNA-seq data we identified a formed using the CFX384 Touch Real-Time PCR mined from strain yLM093. Here we discovered number of mating type specific genes that were Detection System (BioRad, 1854-5485) and analy- a missing loxPsym site in the right arm. Thus, differentially regulated in these three strains sis carried using the CFX Manager Software we reintegrated megachunk I into the synVI.H (Fig. 3C and fig. S16). Further, we identified ex- (BioRad). Single primer pairs targeting each strain (yLM091) and verified the sequence across pected differential expression of auxotrophic of two reference genes, UBC6 and TAF10 (20), megachunk I using Sanger sequencing in four markers (e.g., MET15, LYS2)inpoly-synstrains were designed. separate isolates (yLM172 to 175). We discovered when the genotype did not match BY4741. The r0 10 variant nucleotides on synVI in the genome instances of differential expression provide in- Generating strains of yLM093/yLM175 (table S1). Six of the 10 ternal controls for our RNA-seq analysis pipeline. Strains were grown for 6 hours in synthetic com- mutations preexisted in chunk B3, which had In all poly-syn strains carrying synIII, we ob- plete medium or YPD plus 10-mg/ml EtBr and been previously assembled from three smaller served expression of an allele of ura3.Thisis plated for single colonies on YPD. Inability of sin- fragments prior to digestion and in vitro liga- the result of a mutant URA3 allele at the HO gle colonies to grow on 3% glycerol following rep- tion with B1 and B2 for megachunk incorpora- locus, originally used to select for integration of lica plating was used to infer the loss of the tion. Two alterations occurred at the recoded the essential SYN.SUP61 tRNA.Inbrief,thees- mitochondrial genome. In some cases the loss

BtgI sticky end junction between chunk D2 and sential chromosome III tRNA, SUP61, deleted of the mitochondrial genome further confirmed on March 14, 2017 D3, however these sequence variants corre- during synIII design (6), was integrated into the by PCR, assessing the presence/absence of am- spond to the native genome sequence and thus HO locus on chromosome IV. This integration plicons derived from three loci in the mitochon- arelikelytobeaproductof“patchwork” homol- was achieved with a URA3 marker (6). While at- drial genome (15S rRNA, ATP6, COX2). ogous recombination (6). The remaining two tempting to completely delete the URA3 selec- mutations that give rise to nonsynonymous tion marker with a spanning PCR product, a Correcting two nonsynonymous coding changes are in RIM15 (YFL033C;S1173P) frameshift mutation was introduced during selec- mutations in MOB2 and MSH4 (YFL003C; Y643H) and are presumed tiononFOAratherthandeletionofURA3,yielding The essential MOB2 gene encodes an activator to have occurred spontaneously on transforma- the FOA+, Ura– synIII isolate, yLM422 (6). The of the Cbk1 kinase and is involved in the RAM tion. The complete genome sequence of the error was propagated through all subsequent (regulation of Ace2 activity and cellular mor- synVI strain encoding corrections to MOB2 poly-syn strains carrying synIII. A new set of poly- phogenesis) signaling network regulating cell and PRE4 was also determined (yLM953) at syn strains correcting this error is available: synIII polarity and morphogenesis (21). We discovered

which point we discovered one new variant synVI (yZY455), synIII synIXR (yZY456), synIII that two nonsynonymous mutations in MOB2 http://science.sciencemag.org/ exclusive to this strain (240434 A to C; IRC7 synVI synIXR (yZY175) (table S5). (T155P and F140L) (table S1), inadvertently in- T149P) (table S1). RNA preparation and subsequent analysis of troduced during incorporation of megachunk B, the data for both synVI and the poly-syn RNA- resulted in a slow growth defect on 3% glycerol Poly-syn strains seq experiment was carried out as described (6). plates at 37°C in the first iteration of the synVI The genome sequence of synIII synVI (yLM890), For experiments involving yLM953, yLM890, strain (yLM175) (fig. S3A). We mapped the growth synIII synIXR (yLM758), synVI synIXR (yLM892), yLM758, yLM892, and yLM896 RNA-Seq library defect back to MOB2 using the following strat- and synIII synVI synIXR (yLM896) were all de- prepsweremadeusingtheIlluminaTruSeqRNA egy. First, a MATa r0 synVI strain (yLM189) de- termined (table S3). sample Prep Kit v2 (RS-122-2002), using 500 ng of rived from yLM175, cured of its mitochondrial Downloaded from total RNA as input, amplified by 12 cycles of PCR, genome by growth in ethidium bromide was RNA-seq and run on an Illumina 2500 (v4 chemistry), as backcrossed to a MATa WT strain (BY4742) (fig. SynVI paired-end 50. S3, B and C). Following sporulation of the het- Four isolates of an early version of synVI (yLM172- erozygous diploid (yLM264), we found the slow 175; encoding uncorrected versions of MOB2 Preparation of RNA for qPCR analysis growth defect on glycerol at 37°C segregated 2:2 and PRE4), which correspond to independent Yeast cultures (5 ml) were grown in YPD over- in all tetrads (yLM190-192 and 200-240), suggest- transformants from integration of megachunk night at 30°C with rotation. 250 ml of cells were ing that mutations in the mitochrondrial genome I, plus a single isolate of a WT strain (BY4741) collected by centrifugation at room temperature of the synVI strain did not underlie the defect (fig. were used for the initial transcriptome analysis at 3000 rpm for 3 min. Cells were washed once S3, D to F). Further, using PCRTag analysis we of synVI. In this version of synVI we consistently with water and RNA was extracted using the found that the defect cosegregated with synthetic observed down-regulation of each of the ter- RNeasy Mini Kit (Qiagen, catalog no. 74106). In PCRTagsinallcases(fig.S3,DtoF)indicating minal genes synVI genes, YFL055W and YFR055W, brief, RNA was extracted using enzymatic cell one or more loci on synVI were likely the cause as well as a gene called HAC1 (YFL031W), involved lysis with zymolyase and RNA was isolated as of the phenotype. We exploited meiotic recom- in the unfolded protein response (fig. S5). per the manufacturer’s instructions. DNase treat- bination crossover events between synVI and A second round of RNA-seq analysis using a ment was performed on eluted RNA samples native VI, tracked via presence or absence of syn- “finished” version of synVI (yLM953; encoding (Qiagen, 79254), and the samples were subse- thetic PCRtags, to map the defect to a 20-kb in- corrected versions of MOB2 and PRE4)showed quently purified over a second column to re- terval encoding seven genes within chunk B3 only down-regulation of the terminal genes and move the DNase. The absence of genomic DNA (fig. S3, D to G). The B3 synthetic segment en- HAC1 expression was unchanged (Fig. 1C). This was verified by qPCR using control gene primers coded four non-synonymous mutations (two each result was confirmed by qPCR. Down-regulation for all samples prior to reverse transcriptase treat- in MOB2 and RIM15) (table S1). Additionally, an

Mitchell et al., Science 355, eaaf4831 (2017) 10 March 2017 6of11 RESEARCH | RESEARCH ARTICLE | SYNTHETIC YEAST GENOME intron in each of MOB2 and RPL22B was de- one product generated from the B3 construct were serially diluted in 10-fold increments in leted during design. Although neither deletion (pAW006 with two variant nucleotides; second water and plated onto each type of medium (fig. of RPL22B (yLM305) nor RIM15 (yLM548), both exon, 3′UTR). Here, the construct was also flanked S8). All drugs [methane methanosulfate (MMS) nonessential genes, phenocopied the defect in an by inward-facing BsaI sites encoded by the pri- (Sigma, 129925), Benomyl (Aldrich, 381586), camp- otherwise WT background, deletion of the es- mers, subcloned into pCR-BluntII-TOPO, and tothecin (Sigma, C9911), hydroxyurea (HU) (Sigma, sential WT.MOB2 gene from native chromosome sequence-verified (pLM279). The syn.MOB2 al- H8627)] were mixed into rich medium (YPD) sixphenocopiedtheslowgrowthdefectonglyc- lele encoding both the intron and the repaired except 6-azauracil (Sigma, A1757), which was erol at 37°C in a heterozygous diploid VI/synVI nucleotides (syn.MOB2.intron.2nt) was con- mixed in synthetic complete medium contain- strain (yLM288) (fig. S4A). Consistent with this structed and sequence-verified as described for ing dextrose. YPGE was prepared with 2% gly- observation, installing a WT allele of MOB2 into syn.MOB2.2nt except pAW006 was used as the ercol and 2% ethanol. High- and low-pH plates an otherwise synVI haploid strain rescued the template for the part of the construct encoding (9.0 and 4.0) were prepared using NaOH and glycerol growth defect (yLM321) (fig. S4B). To the second exon (pLM283). In all three cases, BsaI HCl respectively in YPD. Sorbitol plates were test whether the rescue required reintroduc- digestion liberated each syn.MOB2 allele for prepared by adding the appropriate quantity of tion of the MOB2 intron, repair of the two non- integration into a synVI/synVI.syn.mob2D::URA3 sorbitol (Sigma, S1876). For hydrogen peroxide synonymous mutations, or both, all three alleles (yLM294) by homologous recombination. Trans- (Fluka, 88597) and cycloheximide (Sigma C7698), were constructed and introduced into the MOB2 formants were selected on 5-FOA and integra- overnight cultures were treated for 2 hours in locus on synVI. In haploid cells, we found that tions confirmed by PCR. For syn.MOB2.intron drug, harvested by centrifugation and resus- correcting the two nonsynonymous mutations and syn.MOB2.intron.2nt,anintronspecificprim- pended in water prior to plating the serial di- (yLM402) was sufficient to restore the fitness of er was paired with a primer upstream of the lutions on YPD. Plates were incubated at 30°C synVI cells growing on glycerol at 37°C back to integration site. For syn.MOB2.2nt, the reverse unless otherwise indicated in fig. S8 for 2 days that observed in the wild type (fig. S4C). Re- synthetic MOB2 PCRTag was paired with a prim- (YPD, camptothecin, 6-azauracil, Benomyl, pH4.0, turning the intron in combination with correct- er whose 3′ terminal nucleotide was either the pH9.0, cycloheximide, hydrogen peroxide) or ing the coding sequence produced no additive designed or the mutant nucleotide and PCR con- 3 days (sorbitol, YPGE, HU, MMS). effect (yLM366) and returning the MOB2 in- ditions optimized for specific amplification (Tm = tron (yLM383) alone did not rescue the defect 62.5°C). Following sporulation and dissection, in SynVI backcross to map glycerol- (fig. S4C). all cases the same PCR confirmation primers were negative growth-suppression defect

used to assess syn.MOB2 allele segregation. to megachunk H on March 14, 2017 MOB2 strain construction The synVI strain with a corrected MOB2 allele SynVI (yLM175) was crossed to BY472 and hetero- PCR to test for presence of the (yLM402) displayed the glycerol-negative growth- zygous diploids (synVI/VI) selected on medium mitochondrial genome suppression defect. To map the defect, two lacking methionine and lysine (yLM285). MOB2 ThesynVIstrain(yLM402)wasgrowntosatura- spontaneously generated synVI (yLM402) glycerol- was targeted for deletion with a URA3 cassette tion in liquid culture in rich medium (YPD) over- negative colonies were mated to BY4742 and by homologous recombination using standard night at 30°C with rotation and plated for single diploids selected on synthetic medium lacking lithium acetate transformation. Eight transfor- colonies on YPD plates. Following incubation lysine and methionine (yLM578 and yLM579). mants were tested to determine whether the syn- at 30°C for 5 days, the colonies growing on YPD Following sporulation and dissection, eight tet- thetic (SYN.MOB2)ortheWT(WT.MOB2)allele plates were transferred to 3% glycerol plates by rads derived from each parent strain (yLM578 = was deleted based on loss of the amplicon using replica plating. gDNA from 16 glycerol minus tetrads 1 to 8; yLM579 = tetrads 9 to 16 in fig. S9) either the MOB2 synthetic or WT PCRTag primer colonies was prepared (19), along with gDNA were subjected to PCRTag analysis using fifteen + 0 pairs (yLM286 syn.mob2DURA3/WT.MOB2;yLM288 from r and r WT strain (BY4741 with mito- synthetic PCRTags spaced evenly along the length http://science.sciencemag.org/ SYN.MOB2/wt.mob2DURA3). A MOB2/mob2D::URA3 chondrial gDNA intact and BY4741 cured of the of the chromosome and further assessed which of genotype was also constructed in the BY4743 mitochondrial gDNA by growth in ethidium the 64 spores had inherited the glycerol-negative background (yLM290) as well as a synVI/synVI bromide). PCR using three primer pairs target- growth-suppression defect (Fig. 1F and fig. S9). background (yLM294) generated by conditional ing genes in the mitochondrial genome (15S RNA, In all cases the defect segregated 2:2 with the chromosome destabilization (see below); here ATP6, COX2), plus one primer pair targeting a right arm of synVI. The entire set of 164 synthetic deletion of a MOB2 allele in all strains was val- nuclear gene (PRE4) as a positive control was PCRTag amplicons was then used to test two idated by standard PCR confirmation. Sporula- performed and products separated by 1.2% agar- spores, 10B and 13A, for their synthetic DNA tion and dissection of all heterozygous diploid ose gel electrophoresis in TTE (fig. S6). content (fig. S10A). As a result, the defect was Downloaded from MOB2/mob2D::URA3 strains tested yielded 2:2 mapped to a ~30-kb region encompassing mega- dead:alive spore ratios, where all survivors were RIM15 and MSH4 chunk H (YFR034C to YFR053C)(fig.S10B). Ura-, suggesting that MOB2 is essential, contrary Both RIM15 and MSH4 were found to encode to some reports (21, 22). WT and synthetic MOB2 nonsynonymous mutations compared to the de- tK(CUU)F and YFR045W intron alleles were amplified from BY4741 and synVI signed sequence in living synVI strains (yLM175, Two genes in the glycerol-negative growth- (yLM175) genomic DNA (gDNA), respectively, and yLM402, yLM953) (table S1). Each of these two suppression defect mapped interval had annota- used to replace the URA3 cassette in the ap- genes were individually deleted with a URA3 cas- tions relevant to mitochondrial function: deletion propriate heterozygous diploid strains by select- sette in both the WT (BY4741) and synVI (yLM402) of a tRNA [tK(CUU)F] that is imported into the ing for transformants resistant to 5-FOA. All strain backgrounds to generate yLM548, yLM554, mitochondria (23), and deletion of the intron integrations were confirmed by PCR. yLM545, and yLM551. The deletions were con- in YFR045W, encoding a putative mitochon- The syn.MOB2 allele encoding the repaired firmed by PCR. The deletion mutants were used drial import protein. tK(CUU)F was targeted nucleotide sequence (syn.MOB2.2nt) was PCR to investigate whether the nonsynonymous mu- for deletion with a URA3 cassette by homolo- amplified from pJS183 using primers encoding tations were involved in the glycerol-negative gous recombination in BY4741 using standard 5′ inward-facing BsaI sites and subcloned into growth-suppression defect (fig. S7). homologous recombination techniques to gen- the pCR-BluntII-TOPO vector (Invitrogen; cat no erate tK(CUU)FD::URA3 (yLM748). Similarly, 45-0245, Carlsbad, CA) (pLM281). The syn.MOB2 SynVI and IXL-synIXR growth on YFR045W was targeted for deletion with a URA3 allele encoding the intron (syn.MOB2.intron)was different media and conditions cassette by homologous recombination in both constructed by fusion PCR using one product Wild-type (VI and IX, BY4741), synVI (yLM402), BY4741 and synVI (yLM402) using standard ho- generated from BY4741 gDNA (promoter, first and IXL-synIXR (yLM461) strains were grown to mologous recombination techniques to generate exon, intron, and 40 bp of the second exon) and saturation in YPD at 30°C with rotation. Cultures yfr045wD::URA3 (yLM698 and yLM699, respectively).

Mitchell et al., Science 355, eaaf4831 (2017) 10 March 2017 7of11 RESEARCH | RESEARCH ARTICLE | SYNTHETIC YEAST GENOME

The WT.YFR045W and SYN.YFR045W alleles were CEN/ARS/HIS3 fragments were gel-purified. The incubated at 30°C with rotation for 18 hours. amplified from the appropriate gDNA (BY4741 or corresponding linearized chunk and CEN/ARS/ 200 ml of culture was spread evenly onto a plate yLM402, respectively) and the resulting PCR pro- HIS3 fragments were cotransformed into yeast with synthetic complete medium supplemented ducts transformed into the appropriate yfr045wD:: for assembly by homologous recombination and with 5-FOA and extra uracil and after incuba- URA3 strains (WT.YFR045W into yLM699 and selection carried out on synthetic medium lack- tion at 30°C for 2 days, eight single colonies SYN.YFR045W into yLM698) and selection achieved ing histidine (SC–His). For each of the three assem- were streak purified on the same type of medium. on 5-FOA. Integrations were confirmed by PCR. blies, yeast transformants were pooled and plasmids The presence of synthetic DNA and absence of Neither deletion of the tRNA nor intron in a WT recovered into E. coli (Top10 cells) as previously WT DNA was assessed by PCRTagging with ~15 background provoked a glycerol-negative growth- described except the blue-white screening step synthetic and WT PCRTags, evenly spaced along suppression defect (fig. S11). was omitted (25). A correctly assembled clone was the length of chromosome six. A single clone identified by analyzing the restriction digestion was subsequently tested using the entire panel Circularization of synVI and native VI pattern using three sepate enzymes (AflII, BsiWI, of synthetic and WT PCRTags to verify loss of The URA3 gene was amplified with primers en- and XcmI) and subsequently transformed into the full-length WT chromosome (i.e., 2N-1). To coding ~40-bp overhangs targeting the two ends both synVI and WT yeast cells in addition to an determine whether the synVI chromosome had of synVI, just inside the core X elements, for empty vector. The H2, H3, and H4 chunk constructs endoreduplicated, four single colonies derived homologous recombination to generate a ring converted for expression in yeast are stored as from the 2N-1 strain were subjected to sporula- chromosome VI structure (native VI or synVI) pLM356, pLM357, and pLM370, respectively. tion and tetrad dissection. In all four cases the (fig. S12A). A version of the URA3 gene that could To evaluate complementation, three indepen- growth of almost entirely four spore tetrads was enable downstream linearization via the telom- dent transformants were inoculated into SC– consistent with endoreduplication. The homo- erator was used (24). Strains with the resultant His liquid medium and incubated with rotation zygous diploid synVI strain was named yLM544. native VI and synVI ring chromosomes were overnight at 30°C. Single colonies from each OnecopyofPRE4 was targeted for deletion with identified by PCR with primers spanning the strain were then plated on SC–His plates and a URA3 cassette by homologous recombination URA3 gene and the absence of the native VI incubated for 4 days at 30°C. The growth defect using standard lithium acetate transformation in or synVI band tested by pulsed-field gel (ring was measured as a percent of the petite colonies the WT VI/VI (BY4743) and synVI/synVI (yLM544) chromosomes do not penetrate pulsed-field gels) compared to the total number of colonies grow- diploid strains and the deletion confirmed by (Fig. 1B and fig. S12B; yLM717 and yLM728). Of ing (fig. S13). We observed that the petite fre- PCR (yLM770 and yLM772, respectively). A plas- the recovered ring synVI transformants, a sin- quencywasmuchlessprominentinthesynVI mid encoding WT PRE4 ([WT.PRE], pLM394 as on March 14, 2017 gle isolate lacked the glycerol-negative growth- strain on SC–His medium (~15% rather than ~50% described above) was transformed into yLM770 suppression defect, whereby the colony size and on rich medium). When expressed in the synVI and yLM772 and both strains were subjected to morphology matched that of native VI strains strain only chunk H3 rescued the defect (Fig. 2E). sporulation and tetrad dissection. Spores that (circular or linear) (fig. S12C). Pulsed-field gel anal- were Ura+, G418r and had also inherited the ysis revealed a structural rearrangement within Cloning YFR047C, YFR048W, YFR049W, classic MATa and MATa genotype configurations the synVI chromosome in this strain resulting YFR050C, YFR051C, and YFR052W associated with “Boeke Yeast” (BY) strains BY4741 in a linear chromosome ~360 kb long. Genome Each synthetic gene encoded within H3 was and BY4742 (28)(MATa LYS2 met15D;MATa sequencing of the “triplication strain” (yLM644) subsequently cloned individually for expression lys2D MET15) were selected (synVI pre4D::URA3 revealed the rightmost 60 kb of the synthetic in yeast. Primers that annealed ~500 bp upstream [WT.PRE4] yLM823; VI pre4D::URA3 [WT.PRE4] chromosomewaspresentinthreecopiesbased of the start codon and ~200 bp downstream of yLM819). Alleles of PRE4, constructed as de- on sequence read depth (fig. S12D). That triplica- the stop codon were designed with ~40-bp over- scribed below, were integrated directly into the

tion of the right arm of synVI rescues the glycerol- hang sequence homologous to the terminal ends pre4D::URA3 locus by selection on 5-FOA in one of http://science.sciencemag.org/ negative growth-suppression defect suggests that of a BsaI-digested custom CEN/ARS shuttle vec- two ways: (i) DNA was transformed into the het- increased copy number of one or more genes con- tor marked with KanMX (pLM200). Each gene erozygous diploid strains (yLM770 and yLM772) tained in this segment underlies the defect. was amplified from synVI yeast genomic DNA and following sporulationanddissection,spores and assembled into the linearized pLM200 vec- carrying the PRE4 allele of interest were iden- Converting synVI chunks to evaluate tor by one-step isothermal assembly (26) and tified by PCR; or (ii) DNA was transformed into glycerol-negative growth-suppression transformed into E. coli (Top10). Clones encod- the haploid pre4D strains expressing PRE4 epi- defect complementation in yeast ing the correctly sized insert were identified by somally(yLM819andyLM823)andfollowingse- The glycerol-negative growth-suppression defect restriction digestion from plasmid DNA pre- lectionon5-FOAthelossoftheWT.PRE4 plasmid Downloaded from mapped to a ~30-kb interval on the right arm of pared from overnight cultures grown in LB sup- confirmed by replica plating and identifying G418 synVI encoding 23 genes, encompassing YFR034C plemented with carbenicillin (75 mg/ml). [YFR047C], sensitive colonies. In all cases, the correct integra- to YFR053C, spanning chunks H2, H3, and H4. To [YFR048W], [YFR049W], [YFR050C], [YFR051C], tion was identified by PCR. Thus for all strains investigate whether increased copy number of any and [YFR052W]arestoredaspLM364,pLM365, generated to study the relationship between PRE4 DNA in these chunks could rescue the glycerol- pLM367, pLM368, pLM370, and pLM385, respec- and the glycerol-negative growth-suppression de- negative growth-suppression defect, we converted tively. We found that increased copy number of fect, PRE4 was integrated into the precise chromo- each chunk construct for replication and segrega- YFR050C,encodingPRE4,couldrescuethede- somal location and expressed under the control tion in yeast by introducing a centromere (CEN), fect (fig. S13). of the PRE4 promoter, including the N-terminally autonomously replicating sequence (ARS), and tagged HA-PRE4 strains. selectable marker (HIS3)(fig.S13). PRE4 strain construction After integration of SYN.PRE4, WT.PRE4, SYN- A distinct restriction site flanking the synVI PRE4 is an essential gene. A synVI homozygous WT.PRE4,andWT-SYN.PRE4 alleles into yLM823, chunk insert was identified and used to linearize diploid strain was generated by mating a r0 each strain was “cured” of its mitochondrial genomic each of H2 (pLM173, BlpI), H3 (pLM174, NotI) synVI cell (yLM450) to a WT chromosome VI DNAbygrowthinethidiumbromide(describeabove). and H4 (pLM175, BamHI) chunks. Following agar- strain encoding the pGAL1-CEN6::URA3(Kl) con- A r0 isolate was then crossed to a native VI pGAL- ose gel electrophoresis (0.8%), linearized chunks truct (yLM245) (27) in place of the native chro- CEN6 strain(yLM661)andthenativechromo- were gel-purified. Primers were designed to an- mosome VI centromere (yLM661). Diploids were some VI destabilized in the resulting diploid strain neal on either side of the CEN/ARS/HIS3 segment selected by streak purifying on SC medium lack- by growth in galactose (described below). Follow- of pRS413 (16), and additionally encode ~40 bp ing methionine, uracil, and lysine. A single col- ing sporulation and dissection haploid strains with of overhang sequence to the terminal ends of the ony was inoculated into yeast extract/peptone/ “clean” mitochondrial gDNA populations were iso- linearized chunk sequences. Following PCR, the galactose (YP galactose) liquid medium and lated (synVI SYN.PRE4,yLM945;synVIWT.PRE4,

Mitchell et al., Science 355, eaaf4831 (2017) 10 March 2017 8of11 RESEARCH | RESEARCH ARTICLE | SYNTHETIC YEAST GENOME yLM949; synVI SYN-WT.PRE4, yLM953; synVI yLM1020; codon 10, yLM1022). Integrations were 180 mg of peptides from each sample were labeled WT-SYN.PRE4,yLM957).Followingthis“endor- confirmed by PCR and single codon changes with TMT reagent according to manufacturer’s eduplication backcross” to introduce WT pop- validated by sequencing. protocol (Thermo Scientific, 90110). 20 mgofpep- ulations of mitochondria, the glycerol-negative The SYN.PRE4.WT.RSSR and WT.PRE4.SYN.RSSR tides from each of the 9 individual samples were growth-suppression defect was not cured in the alleles were also generated by fusion PCR, where- combined and used as a universal standard. The strains encoding the SYN.PRE4 allele. by the four codons to be changed were encoded TMT reagents (0.8 mg) (Thermo Scientific) were within overlapping central primers. The 5′ and dissolved in 44 ml of ACN, 10 ml was added to Constructing PRE4 alleles 3′ PCR products were subjected to fusion PCR each sample along with 20 mlofACN.Afterin- WT.PRE4 and SYN.PRE4 alleles were generated with the two flanking primers and products were cubatingfor1houratroomtemperature(25°C), by PCR using genomic DNA extracted from WT subcloned into the pCR-BluntII-TOPO vector the reaction was quenched with 4 mlof5%w/v (BY4741) and synVI (yLM402) cells as template, (Invitrogen; 45-0245) and sequence-verified (SYN. hydroxylamine. Following labeling, the samples respectively. The primers annealed ~500 bp up- PRE4.WT.RSSR, pLM444; WT.PRE4.SYN.RSSR, were combined in equal ratio. The pooled sam- stream and ~200 bp downstream of the PRE4 pLM445). For integration into yLM823 and yLM819 ple was desalted over SCX and SAX solid-phase start and stop codons to provide template for ho- the alleles were excised from the plasmid back- extraction columns (SPE, STRATA, Phenomenex). mologous recombinationupontransformation bonesbydoubledigestionwithBsaIandSspI. into yeast cells. The PCR products were trans- Offline basic-pH RP fractionation formed into heterozygous diploid PRE4/preD SynIII synVI synIXR proteome analysis The pooled TMT labeled sample was subjected strains encoding two copies of native chromo- Cell growth to basic-pH reverse-phase HPLC for fractionation. some VI (yLM770) or two copies of synVI (yLM772) BY4741 and synIII synVI synIXR cultures (200 ml), The mixture was solubilized in buffer A (10 mM with selection on 5-FOA. After confirming in- transformed with a plasmid encoding a kanMX ammonium formate, pH 10.0). Basic pH HPLC tegration by PCR, the strains were subjected to cassette (pLM200), were grown in YPD + G418 was performed on a 4.6 mm × 250 mm Xbridge sporulation and dissection to generate WT hap- at 30°C to an OD of 1.0. Cell pellets were col- C18 column (Waters, 3.5 mm bead size) using an loid cells carrying the synthetic PRE4 allele (VI lected by centrifugation (3000 rpm, 3 min) at 4°C Agilent 1260 HPLC instrument. A 30-min linear SYN.PRE4, yLM849) or synVI cells carrying the and washed once with ice cold water and subse- gradient from 10 to 50% solvent B (90% ACN, WT PRE4 allele (synVI WT.PRE4, yLM848). Segre- quently frozen at –80°C. 10 mM ammonium formate, pH 10.0) at a flow gation of the synthetic and WT alleles in spores rate of 0.5 ml/min separated the peptide mixture wastestedbyPCRTagprimersandinallcases Cell lysis into a total of 30 fractions. The 30 fractions were on March 14, 2017 segregated two to two. Cells were lysed in 15-ml centrifuge tubes in TMT pooled into 10 final fractions by combining early, WT-SYN.PRE4 and SYN-WT.PRE4 alleles were lysis buffer [10 mM Tris/HCl (pH 8.0), 8 M urea, middle, and late eluting RPLC fractions. The constructed by separately amplifying the 5′ and 0.15% deoxycholate] plus protease inhibitor (Roche, concatenated samples were then dried in the 3′ halves of each of SYN.PRE4 and WT.PRE4, 11873580001, added just prior to lysis) and 0.5-mm Speedvac and stored at –80°C till LC-MS/MS splitting the gene between the two PCRTags. The glass beads (10:1 ratio of lysis buffer to cells, 2:1 analysis. PCR amplicons encoded 40-bp terminal sequence volume of cell slurry to beads). Lysis was achieved homology to facilitate fusion PCR to generate WT- in 10 × 30 s vortex blasts separated by 1 min of LC-MS/MS SYN.PRE4 and SYN-WT.PRE4 alleles by pairing the rest on ice followed by 3 × 20 s sonication steps Concatenated samples were analyzed with the appropriate PCR products. Fusion PCR amplicons (Sonicator Xx, setting 4). Sonication was per- Q-Exactive mass spectrometer (Thermo Scien- were subcloned into the pCR-BluntII-TOPO vec- formed at 4°C with 2 min of recovery time be- tific) coupled to an EASY-nLC (Thermo Scien- tor (Invitrogen; 45-0245) and sequence-verified tween pulses. tific) liquid chromatography system. The sample

(pLM433 WT-SYN.PRE4;pLM435SYN-WT.PRE4). was first loaded onto a trap column (Acclaim http://science.sciencemag.org/ For integration into yLM823 and yLM819 the Protein digestion PepMap 100 pre-column, 75 mm×2cm,C18,3mm, alleles were excised from the plasmid backbones Lysate protein concentration was measured at 100 Å, Thermo Scientific) and then separated on by double digestion with BsaIandSspI. 280 nm using a NanoDrop DS-11 Spectropho- an analytical column (EASY-Spray column, 50 cm × N-terminally tagged 3HA PRE4 alleles were gen- tometer (DeNovix). Disulfide bonds were re- 75 mmID,PepMapRSLCC18,2mm, 100 Å, Ther- erated by fusion PCR of three separate amplicons: duced by addition of 5 mM dithiothreitol (DTT) mo Scientific) maintained at a constant temper- (i) ~300 bp upstream of the PRE4 ATG, (ii) 3HA and incubated for 1 hour at 55°C. The mixture ature (45°C) and equilibrated with 5% buffer A tag, which was ordered as an ultramer from In- was cooled to room temperature, and the reduced (2% acetonitrile, 0.5% acetic acid). Peptides were tegrated DNA Technologies (IDT), and (iii) the cysteine residues were alkylated by addition of separated with a 120-min linear gradient of 5 to Downloaded from PRE4 coding sequence plus ~200 bp downstream 14 mM iodoacetamide and incubation in the 30%bufferB(90%acetonitrile,0.5%aceticacid) of the stop codon. The fusion PCR product was dark at room temperature for 45 min. Excess at a flow rate of 200 nl/min. Each full MS scan subcloned into the pCR-BluntII-TOPO vector iodoacetamide was quenched with an additional [Resolution (R) = 70,000 @ m/z 200] was fol- (Invitrogen; 45-0245) and sequence-verified (HA- 5 mM dithiothreitol. The cell lysate was digested lowed by data dependent MS2 scan (R = 35,000 @ WT.PRE4 pLM437; HA-SYN.PRE4 pLM432). For with Lys-C (Promega, 200:1 by weight) at room m/z 200) with HCD and an isolation window of integration into yLM823 and yLM819 the alleles temperature for 2 hours in lysis buffer (8 M 1.6 m/z. Normalized collision energy was set to were excised from the plasmid backbones by urea, 10 mM Tris-HCl, pH 8.5), then the urea 27, target value of 1e6 for MS1 and 5 × 104 for double digestion with BsaIandSspI. concentration was diluted to 2 M by addition MS2, maximum ion time 50 and 180 ms for MS1 To construct PRE4 alleles in which individual of 10 mM Tris-HCl, pH 8.5 and the sample di- and MS2, respectively, monoisotopic precursor synthetic codons were replaced with the WT gested with Trypsin (Promega, 100:1 by weight) selection was enabled and the dynamic exclu- versions (across the 3′ PCRTag), PCR products at room temperature overnight. The sample was sion was set to 15.0 s. were generated corresponding to the 5′ and 3′ acidified with TFA to pH < 3 and desalted using halves of PRE4, splitting the gene within the 3′ C18 solid-phase extraction (SPE, Sep-Pak, Waters), Data analysis PCRTag and encoding the WT codon in the over- andelutedwith80%acetonitrile(ACN)in0.5% Raw mass spectrometry data were processed lapping primer. Fusion PCR products, generated acetic acid. The eluent was dried in the Speedvac using MaxQuant version 1.5.2.8, with a false- with the flanking primer pairs, were used direct- and stored at –80°C until further analysis. discovery rate (FDR) of 0.01 at the level of pro- ly for yeast transformation (codon 1, yLM1004; teins and peptides, using the following settings: codon 2, yLM1006; codon 3, yLM1008; codon 4, TMT labeling oxidized methionine (M), acetylation (protein yLM1010; codon 5, yLM1012; codon 6, yLM1014; The desalted peptides were re-suspended in N-term) and deamidation (NQ) were selected codon 7, yLM1016; codon 8, yLM1018; codon 9, 70 ml of 50 mM HEPES (pH 8.5). Approximately as variable modifications, and carbamidomethyl

Mitchell et al., Science 355, eaaf4831 (2017) 10 March 2017 9of11 RESEARCH | RESEARCH ARTICLE | SYNTHETIC YEAST GENOME

(C) as fixed modifications; precursor mass toler- tor (Invitrogen; 45-0245) and sequence-verified 2. J. S. Dymond et al., Synthetic chromosome arms function ance 10 ppm; fragment mass tolerance 0.01 Th; (SYN.HIS2-HA, pLM532; WT.HIS2-HA,pLM533). in yeast and generate phenotypic diversity by design. Nature – and minimum peptide length of six amino acids. For integration into yLM1130 and yLM1131 the 477, 471 476 (2011). doi: 10.1038/nature10403; pmid: 21918511 Proteins and peptides were identified using a alleles were excised from the plasmid backbones 3. R. E. Chapman, P. Walter, Translational attenuation target-decoy approach with a reversed database, by digestion with BsaI. The digestion product was mediated by an mRNA intron. Curr. Biol. 7, 850–859 (1997). using the Andromeda search engine integrated transformed and following selection on 5-FOA, doi: 10.1016/S0960-9822(06)00373-3; pmid: 9382810 4. U. Rüegsegger, J. H. Leber, P. Walter, Block of HAC1 mRNA into the MaxQuant environment. Searches were transformants confirmed by PCR (chrVI WT.HIS2- translation by long-range base pairing is released by performed against the UniProt S. cerevisiae FASTA HA, yLM1147; chrVI SYN.HIS2-HA, yLM1150, synVI cytoplasmic splicing upon induction of the unfolded database. The following filters and criteria were WT.HIS2-HA, yLM1153; synVI SYN.HIS2-HA, protein response. Cell 107, 103–114 (2001). doi: 10.1016/ used for quantification: (i) Only distinctive pep- yLM1156). To modify the region 5′ of HIS2,encod- S0092-8674(01)00505-0; pmid: 11595189 5. See Materials and methods section. tides were used for quantification, and only pro- ing either a tRNA gene (tA(AGC)F) in the WT 6. N. Annaluru et al., Total synthesis of a functional designer teins with at least 2 distinctive peptides were strain or a loxPsym site in the synVI strain, a eukaryotic chromosome. Science 344,55–58 (2014). quantified. (ii) The minimum reporter ion inten- URA3 cassette targeting deletion of these fea- doi: 10.1126/science.1249252; pmid: 24674868 sity was set to 1000. Bioinformatics analysis was tures was transformed into yLM1147, yLM1150, 7. J. L. Schreve, J. M. Garrett, Yeast Agp2p and Agp3p function as amino acid permeases in poor nutrient conditions. performed with Perseus, Microsoft Excel and R yLM1153, and yLM1156 and confirmed by PCR Biochem. Biophys. Res. Commun. 313, 745–751 (2004). statistical computing software. Each reporter ion in all cases. PCR products spanning the tRNA or doi: 10.1016/j.bbrc.2003.11.172; pmid: 14697254 channel was summed across all quantified pro- the loxPsym site (200 bp of flanking homology) 8. M. Roncoroni et al., The yeast IRC7 gene encodes a b-lyase responsible for production of the varietal thiol 4-mercapto-4- teins and normalized assuming equal protein derived from WT (BY4741) or synVI (yLM953) cells methylpentan-2-one in wine. Food Microbiol. 28, 926–935 loading across all ten samples. Welch’s t test was were then used to overwrite the URA3 sequence by (2011). doi: 10.1016/j.fm.2011.01.002; pmid: 21569935 then used to identify proteins that were differ- selectionon5-FOA(chrVIloxPsym WT.HIS2-HA, 9. G. Fourel, E. Revardel, C. E. Koering, E. Gilson, Cohabitation entially expressed across each sample triplicates, yLM1199; chrVI loxPsym SYN.HIS2-HA,yLM1201; of insulators and silencing elements in yeast subtelomeric regions. EMBO J. 18, 2522–2537 (1999). doi: 10.1093/emboj/ and the method of permutation-based FDR (29) synVI tRNA WT.HIS2-HA,yLM1203;synVItRNA 18.9.2522; pmid: 10228166 was subsequently applied to control for multiple SYN.HIS2-HA,yLM1205). 10. W. Hilt, C. Enenkel, A. Gruhler, T. Singer, D. H. Wolf, The PRE4 testing error. gene codes for a subunit of the yeast proteasome necessary Immunoblot analysis for peptidylglutamyl-peptide-hydrolyzing activity. Mutations Direct testing of protein levels in synIII link the proteasome to stress- and ubiquitin-dependent For both HA-Pre4, evaluation of TMT proteome re- proteolysis. J. Biol. Chem. 268, 3479–3486 (1993). synVI synIXR sults, and His2-HA experiments, cells were grown pmid: 8381431 on March 14, 2017 From the list of eleven proteins with altered ex- to mid-log phase at 30°C in YPD medium, col- 11. P. C. Ramos, A. J. Marques, M. K. London, R. J. Dohmen, b b pression level detected in the TMT proteome anal- lected by centrifugation, washed once in ice cold Role of C-terminal extensions of subunits 2 and 7in assembly and activity of eukaryotic proteasomes. J. Biol. Chem. ysis of synIII synVI synIXR, six were selected water, and immediately frozen at –80°C. Cells were 279,14323–14330 (2004). doi: 10.1074/jbc.M308757200; for direct testing by Western blot analysis. His2 lysedinanequalvolumeoflysisbuffer[20mM pmid: 14722099 12. M. Groll et al., Structure of 20S proteasome from yeast at was chosen because it is encoded by synVI. Sac7, HEPES, pH 7.4, 0.1% Tween 20, 2 mM MgCl2, 2.4 Å resolution. Nature 386, 463–471 (1997). doi: 10.1038/ Snu114, Has1, and Chs2 were chosen because they 300 mM NaCl, protease inhibitor cocktail (Roche, 386463a0; pmid: 9087403 are encoded by essential genes. Cox5A was chosen 11873580001)] in the presence of 0.5 mm glass 13. W. Heinemeyer, M. Fischer, T. Krimmer, U. Stachon, D. H. Wolf, because it is an inner mitochondrial membrane beads (1:1 cell slurry:beads) by vortexing (7 × 1 min The active sites of the eukaryotic 20 S proteasome and their protein. Each gene was C-terminally tagged with blastswith1-minincubationonicebetweeneach involvement in subunit precursor processing. J. Biol. Chem. 272, 25200–25209 (1997). doi: 10.1074/jbc.272.40.25200; a3-HAtagusingHIS3 selection as previously blast). Following centrifugation (14000 rpm, 4°C, pmid: 9312134 described (30)inbothWTcells(BY4742)andthe 10 min), 30 ml of clarified whole-cell extract was 14. K. M. Chan, Y. T. Liu, C. H. Ma, M. Jayaram, S. Sau, The 2 Saccharomyces cerevisiae synIII synVI synIXR strain (yLM896). All strains mixed with 10 ml4XLDSsamplebuffer(pre- micron plasmid of : A miniaturized http://science.sciencemag.org/ were confirmed by PCR (yLM1104-yLM1127). Two mixed with beta mercaptoethanol) (Life Tech- selfish genome with optimized functional competence. Plasmid 70,2–17 (2013). doi: 10.1016/j.plasmid.2013.03.001; independent transformants from each of the WT nologies, NP0007). Samples were heated at 70°C pmid: 23541845 and triple-syn strain backgrounds were used to for 10 min and loaded onto a 4 to 12% pre-cast 15. Y. Wu et al., Bug mapping and fitness testing of chemically assess protein level by immunoblot (fig. S17). Bis-Tris gel (Life Technologies, NP03222) and elec- synthesized chromosome X. Science 355, eaaf4706 (2017). 16. R. S. Sikorski, P. Hieter, A system of shuttle vectors and trophoresed in 1x MES buffer. Protein transfer HIS2-HA (native HIS2 3′ sequence) yeast host strains designed for efficient manipulation of DNA was carried out using a BioRad Trans-blot Turbo in Saccharomyces cerevisiae. Genetics 122,19–27 (1989). strain construction in synVI Transfer system and corresponding reagents. pmid: 2659436 HIS2 was deleted with a URA3 cassette in a WT Anti-HA antibody was from Covance (MMS-101P; 17. R. D. Gietz, Yeast transformation by the LiAc/SS carrier Downloaded from Methods Mol. Biol. – (BY4741) and synVI (yLM953) strain background mouse) and Anti-Tub1 (Rabbit) served as a control DNA/PEG method. 1205,1 12 (2014). doi: 10.1007/978-1-4939-1363-3_1; pmid: 25213235 to generate yLM1130 and yLM1131, respectively. for loading. Secondary antibodies were from LI- 18. M. Lisby, U. H. Mortensen, R. Rothstein, Colocalization of The deletions were confirmed by PCR. Synthetic COR (IRDye 800CW, 926-32210; IRDye 680RD, multiple DNA double-strand breaks at a single Rad52 repair and WT C-terminally tagged HIS2-HA alleles 926-68071). Western blots were developed and centre. Nat. Cell Biol. 5, 572– 577 (2003). doi: 10.1038/ncb997; were generated by fusion PCR of three parts: quantified using the LI-COR Odyssey and Image pmid: 12766777 19. J. S. Dymond, Preparation of genomic DNA from Saccharomyces (i) 200 bp upstream of the start codon to the last Studio Software. cerevisiae. Methods Enzymol. 529,153–160 (2013). doi: 10.1016/ coding amino acid, excluding the stop codon; B978-0-12-418687-3.00012-4; pmid: 24011043 (ii) 3xHA sequence amplified from a preexisting Strains and media 20. M. A. Teste, M. Duquenne, J. M. François, J. L. Parrou, plasmid (30); and (iii) ~200 bp downstream of The list of “final” synthetic strains used in this Validation of reference genes for quantitative expression analysis by real-time RT-PCR in Saccharomyces cerevisiae. the stop codon, excluding the loxPsym site from study is listed in table S5. SynVI strain versions BMC Mol. Biol. 10, 99 (2009). doi: 10.1186/1471-2199-10-99; the synthetic HIS2 allele. Synthetic or WT HIS2 are listed in table S6, including details on muta- pmid: 19874630 sequences were amplified from synVI (yLM953) tions identified on synVI in each strain. All media 21. E. L. Weiss et al., The Saccharomyces cerevisiae Mob2p-Cbk1p or WT (BY741) genomic DNA, respectively. Prim- and growth conditions were standard. 3% glycerol kinase complex promotes polarized growth and acts with ers for the three segments were designed to plates were prepared with 3% glycerol (v/v) as the mitotic exit network to facilitate daughter cell-specific localization of Ace2p transcription factor. J. Cell Biol. yield 20 to 30 bp of terminal homology between the sole carbon source, omitting dextrose entirely 158, 885–900 (2002). doi: 10.1083/jcb.200203094; adjacent parts to facilitate fusion PCR. Further, from the recipe. pmid: 12196508 the right- and leftmost primers encoded inward 22. L. Schneper, A. Krauss, R. Miyamoto, S. Fang, J. R. Broach, facing BsaI site to excise the fragment from a The Ras/protein kinase A pathway acts in parallel with the REFERENCES AND NOTES Mob2/Cbk1 pathway to effect cell cycle progression and vector in subsequent steps. Fusion PCR products 1. S. Richardson et al., Design of a synthetic yeast genome. proper bud site selection. Eukaryot. Cell 3, 108–120 (2004). were subcloned into the pCR-BluntII-TOPO vec- Science 355, 1040–1044 (2017). doi: 10.1128/EC.3.1.108-120.2004; pmid: 14871942

Mitchell et al., Science 355, eaaf4831 (2017) 10 March 2017 10 of 11 RESEARCH | RESEARCH ARTICLE | SYNTHETIC YEAST GENOME

23. R. P. Martin, J. M. Schneller, A. J. Stahl, G. Dirheimer, Import of 30. M. S. Longtine et al., Additional modules for versatile and reviewed and managed by the committees on conflict of interest at nuclear deoxyribonucleic acid coded lysine-accepting transfer economical PCR-based gene deletion and modification NYULMC (J.D.B.) and Johns Hopkins University (J.S.B.). S.R. ribonucleic acid (anticodon C-U-U) into yeast mitochondria. in Saccharomyces cerevisiae. Yeast 14, 953–961 (1998). and J.D.B. designed synVI. H.D. and K.D. are employees Biochemistry 18, 4600–4605 (1979). doi: 10.1021/ doi: 10.1002/(SICI)1097-0061(199807)14:10<953::AID- of GenScript, who constructed most of the synVI synthetic DNA bi00588a021; pmid: 387075 YEA293>3.0.CO;2-U; pmid: 9717241 chunks. L.A.M., Y.Z., and J.A.M. (in the lab of J.D.B.), together 24. L. A. Mitchell, J. D. Boeke, Circular permutation of a synthetic 31. M. Zuker, Mfold web server for nucleic acid folding with A.W. (in the labs of J.D.B. and J.D.) and R.W. and Y.L. (in the lab eukaryotic chromosome with the telomerator. Proc. Natl. and hybridization prediction. Nucleic Acids Res. 31, of Y.C.), performed experiments. A.H. oversaw all RNA and DNA Acad. Sci. U.S.A. 111, 17003–17010 (2014). doi: 10.1073/ 3406–3415 (2003). doi: 10.1093/nar/gkg595; sequencing efforts. Y.Y. (in the lab of B.U.) performed tandem pnas.1414399111; pmid: 25378705 pmid: 12824337 mass tag proteomic experiments and analysis. G.S. and K.Y. (in 25. Q. Lin et al., RADOM, an efficient in vivo method for the lab of J.S.B.), Z.K. (in the lab of J.D.B.), and Z.T. and X.W. (in assembling designed DNA fragments up to 10 kb long in ACKNOWLEDGMENTS the lab of D.F.) performed computational analyses. L.A.M. and J.D.B. Saccharomyces cerevisiae. ACS Synth. Biol. 4, 213–220 This work was supported in part by U.S. NSF grants MCB- wrote the manuscript, and all authors contributed to its editing. All (2015). pmid: 24895839 1026068 and MCB-1158201 to J.D.B. L.A.M. was funded by a genomic data for this paper are available under the Sc2.0 umbrella 26. D. G. Gibson et al., Enzymatic assembly of DNA molecules up postdoctoral fellowship from the Natural Sciences and BioProject accession number PRJNA351844. Proteomic data are to several hundred kilobases. Nat. Methods 6, 343–345 Engineering Research Council of Canada. GenScript funded available through MassIVE with reference number MSV000080584. (2009). doi: 10.1038/nmeth.1318; pmid: 19363495 the synthesis of 150 kb of synVI DNA chunks with corporate Additional information related to synVI [design diagram, PCRTag 27. R. J. Reid et al., Chromosome-scale genetic mapping using a funds. We thank P. Meluh for thoughtful discussions and sequences, feature summary table (WT VI, designed VI, physical strain set of 16 conditionally stable Saccharomyces cerevisiae K. Palacios for analysis of genome sequence data. The New York yLM953; yeast_chr6_9_03), variants in physical strain chromosomes. Genetics 180, 1799–1808 (2008). doi: 10.1534/ University Langone Medical Center (NYULMC) Genome (yeast_chr6_3_09), chunks used for assembly, and an extended yeast genetics.108.087999; pmid: 18832360 Technology Center is partially supported by the Cancer Center strain table] can be accessed on the Sc2.0 website. 28. C. B. Brachmann et al., Designer deletion strains derived from Support Grant (P30CA016087) at the Laura and Isaac Saccharomyces cerevisiae S288C: A useful set of strains and Perlmutter Cancer Center. Work in the United Kingdom was SUPPLEMENTARY MATERIALS plasmids for PCR-mediated gene disruption and other applications. funded by a Chancellor’s Fellowship from the University of www.sciencemag.org/content/355/6329/eaaf4831/suppl/DC1 Yeast 14,115–132 (1998). doi: 10.1002/(SICI)1097-0061 Edinburgh, a start-up fund from Scottish Universities Life Sciences Figs. S1 to S17 (19980130)14:2<115::AID-YEA204>3.0.CO;2-2; pmid: 9483801 Alliance, and UK Biotechnology and Biological Sciences Research Tables S1 to S6 29. S. J. Humphrey, S. B. Azimifar, M. Mann, High-throughput Council grants (BB/M005690/1, BB/M025640/1, and BB/M00029X/ References phosphoproteomics reveals in vivo insulin signaling dynamics. 1) to Y.C. J.D.B. and J.S.B. are founders and directors of Nat. Biotechnol. 33, 990–995 (2015). doi: 10.1038/nbt.3327; Neochromosome Inc. J.D.B. serves as a scientific adviser to 15 February 2016; accepted 1 February 2017 pmid: 26280412 Recombinetics Inc. and Sample6 Inc. These arrangements are 10.1126/science.aaf4831 on March 14, 2017 http://science.sciencemag.org/ Downloaded from

Mitchell et al., Science 355, eaaf4831 (2017) 10 March 2017 11 of 11 Synthesis, debugging, and effects of synthetic chromosome consolidation: synVI and beyond Leslie A. Mitchell, Ann Wang, Giovanni Stracquadanio, Zheng Kuang, Xuya Wang, Kun Yang, Sarah Richardson, J. Andrew Martin, Yu Zhao, Roy Walker, Yisha Luo, Hongjiu Dai, Kang Dong, Zuojian Tang, Yanling Yang, Yizhi Cai, Adriana Heguy, Beatrix Ueberheide, David Fenyö, Junbiao Dai, Joel S. Bader and Jef D. Boeke (March 9, 2017) Science 355 (6329), . [doi: 10.1126/science.aaf4831] Editor's Summary

This copy is for your personal, non-commercial use only. on March 14, 2017 Article Tools Visit the online version of this article to access the personalization and article tools: http://science.sciencemag.org/content/355/6329/eaaf4831 Permissions Obtain information about reproducing this article: http://www.sciencemag.org/about/permissions.dtl http://science.sciencemag.org/ Downloaded from

Science (print ISSN 0036-8075; online ISSN 1095-9203) is published weekly, except the last week in December, by the American Association for the Advancement of Science, 1200 New York Avenue NW, Washington, DC 20005. Copyright 2016 by the American Association for the Advancement of Science; all rights reserved. The title Science is a registered trademark of AAAS.